GLM-4.7-FlashOpen Weight

Released

GLM-4.7-Flash is an AI model released by Z.ai on Monday, Jan 19 2026, 28 days after GLM-4.7. It is an open-weight model — the trained weights are available to download and run. It is a 30B parameterThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. model with a 200k token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover SWE-Bench Verified, LiveCodeBench, BrowseComp, and AIME 2025.

API pricing

Input
$0.00
Cached input
$0.00
Output
$0.00
  • Reasoning or thinking is supported.
USD per 1M tokensVerified November 17, 2026Read from docs.z.ai. Click the check to open the source.

Benchmarks

Coding

SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.
59.2%
#63 of 89
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better.
64%
#34 of 70

Agentic & tool use

BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better.
42.8%
#40 of 45

Reasoning & science

AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools
91.6%
#16 of 36

Compare GLM-4.7-Flash with

GLM-4.7-Flash

About

GLM-4.7-Flash launched in January 2026 as a compact reasoning and coding model. It used a mixture of experts with 30 billion total parameters and about 3 billion active per token, giving developers a smaller deployment option in the GLM-4.7 family.

Z.ai released MIT-licensed weights and offered a free hosted API. The model card reported 59.2% on SWE-bench Verified and 64.0% on LiveCodeBench v6 for the Flash checkpoint.

Frequently asked questions

GLM-4.7-Flash was released by Z.ai on Monday, Jan 19 2026.

All Z.ai releases

24 tracked